Digital Advertising and Media Management

Introduction and context

Digital advertising is the paid communication of a brand, product, service, or social cause through digitally connected media. It includes search advertising, display advertising, online video, native advertising, social advertising, retail media, connected television, mobile advertising, and programmatic buying. Media management is the managerial process through which an organisation selects audiences, channels, placements, timing, budgets, messages, measurement systems, and controls to achieve communication and business objectives.

Digital advertising differs from traditional advertising because exposure can be bought, delivered, measured, optimised, and sometimes personalised in near real time. This flexibility does not make planning unnecessary. On the contrary, the abundance of inventory, data, platforms, formats, and measurement choices makes disciplined planning more important. A media plan must connect business objectives to audience insight, channel roles, buying methods, creative executions, budget allocation, measurement, and governance.

A useful distinction is between paid media, which an organisation purchases; owned media, which it controls, such as its website or email database; and earned media, which results from voluntary sharing, reviews, publicity, or advocacy. Digital advertising is primarily paid media, but effective campaigns usually coordinate all three.

Learning objectives

By the end of this chapter, students should be able to:

1. Explain the role and characteristics of major digital advertising formats.
2. Design a basic media plan using audience targeting, reach, frequency, budget, and channel roles.
3. Explain how programmatic auctions and automated delivery work.
4. Calculate and interpret common media metrics such as CPM, CTR, conversion rate, CPA, and ROAS.
5. Evaluate creative effectiveness, native advertising, video, retail media, brand safety, privacy, and advertising fraud.
6. Make evidence-based optimisation and governance recommendations.

1. The media-planning system

A digital media plan is not merely a list of platforms. It is a structured response to six questions:

• Whom is the organisation trying to influence?
• What behavioural or attitudinal change is required?
• Where can the audience be reached credibly and efficiently?
• When and how often should communication appear?
• With what message and format should the audience be engaged?
• How will success and risk be evaluated?

The planning sequence normally begins with the marketing objective, such as increasing qualified leads, launching a product, generating online sales, or improving consideration. The planner then defines the target audience, identifies the customer journey and relevant moments, chooses channel roles, estimates reach and frequency, allocates budget, prepares creative variants, establishes measurement, and specifies optimisation rules.

Figure: The digital media planning process is iterative rather than strictly linear.

Audience targeting

Targeting may be based on demographic characteristics, geography, device, language, interests, declared preferences, contextual content, purchase intent, first-party customer data, or predictive models. First-party data is collected directly by the organisation, such as customer transactions or consented website behaviour. Second-party data is another organisation’s first-party data shared through an agreement. Third-party data is collected and aggregated by external providers, often across many sites or services.

Targeting should be sufficiently specific to improve relevance but not so narrow that delivery becomes expensive, unrepresentative, or impossible. A planner should distinguish between an audience definition and a targeting mechanism. For example, “potential university students interested in business education” is an audience definition; contextual keywords, geographic boundaries, or a consented customer list are possible mechanisms for reaching it.

Modern privacy restrictions and the decline of unrestricted third-party cookies have increased the importance of first-party data, contextual targeting, clean-room environments, consent management, and aggregated measurement. Personalisation also requires restraint: a message that is technically relevant may feel intrusive if the audience does not understand why it was shown.

Reach, frequency, and effective exposure

Reach is the number or proportion of distinct people exposed to an advertisement at least once during a specified period. Frequency is the average number of exposures among reached people. If total impressions are known, average frequency is:

where F is average frequency, I is total impressions, and R is the number of unique reached individuals. Both I and R must refer to the same campaign, period, and measurement universe. The formula assumes impressions are counted consistently and that reach is deduplicated.

If reach is expressed as a percentage of a defined target population N, then:

These measures do not reveal whether exposure was viewable, noticed, or persuasive. Excessive frequency can cause irritation, creative fatigue, and wasted spend, whereas insufficient frequency may fail to build memory. Frequency caps limit how many times a user can be served an ad within a period, although caps may be imperfect when identity is fragmented across devices and platforms.

2. Digital advertising formats and channel roles

Display advertising

Display advertising includes banner, rich-media, responsive, and interactive units placed on websites, applications, and digital services. Display is useful for reach, remarketing, product explanation, and direct response. Its strengths include scale, flexible creative, and measurable clicks. Its limitations include banner blindness, accidental clicks, low attention, viewability problems, and fraud.

Display planning should specify placement quality, above-the-fold versus below-the-fold inventory, screen environment, viewability requirements, frequency controls, and exclusions. A high click-through rate alone does not prove success; accidental or low-quality clicks can inflate the metric without producing value.

Online video and connected television

Video can create emotional involvement, demonstrate product use, and communicate complex propositions quickly. Formats include skippable in-stream, non-skippable in-stream, out-stream, short-form social video, and connected television. Important decisions include sound-on versus sound-off design, opening-frame clarity, duration, subtitles, sequential storytelling, completion objectives, and the difference between a completed view and meaningful attention.

Connected television often provides television-like reach with digital delivery and audience controls. However, device-level impressions may not equal unique household or person-level reach. Measurement should therefore clarify whether “reach” represents devices, households, logged-in users, or modeled individuals.

Native advertising

Native advertising adopts the visual form and editorial conventions of the surrounding environment. Examples include sponsored articles, recommendation widgets, promoted posts, and sponsored search results. Native formats can reduce interruption and support content-led persuasion, but their commercial nature must remain clear. Disclosure such as “Sponsored” or “Advertisement” should be prominent, understandable, and not hidden through ambiguous labels.

Native advertising is appropriate when the audience needs education, comparison, or contextual explanation. It is less appropriate when the format intentionally disguises persuasion as independent editorial content.

Retail media

Retail media is advertising sold by retailers or commerce platforms using their digital properties, shopper data, search results, product pages, physical stores, or connected media. Sponsored product listings, display placements on retailer websites, and off-site audience extensions are common forms. Retail media is attractive because it can connect exposure with product discovery, basket activity, and transaction data.

Its risks include measurement inconsistency, high dependence on a small number of platforms, limited transparency into auction rules, and possible conflicts between the retailer’s commercial interests and the advertiser’s need for independent evaluation. Retail media should be assessed not only by immediate sales but also by incremental sales, new-customer acquisition, margin, and effects on organic product visibility.

Search and social advertising

Although this chapter emphasises display, video, native, retail media, and programmatic advertising, search and social often provide essential complementary roles. Search captures expressed intent, while social platforms can provide discovery, community-based distribution, creator partnerships, and behavioural signals. Platform metrics should be reconciled with independent analytics wherever possible because each platform may define impressions, views, conversions, and attribution differently.

3. Buying models and programmatic advertising

Digital inventory may be bought through direct negotiation, insertion orders, ad networks, private marketplaces, or open programmatic exchanges. Programmatic advertising uses software, data, and automated decision rules to buy and deliver advertising inventory. Automation improves speed and scale, but it does not remove the need for human strategy, quality controls, contracts, or interpretation.

The main participants are:

• Advertiser: funds the campaign and supplies objectives and creative.
• Agency or trading desk: may plan, buy, and optimise media on the advertiser’s behalf.
• Demand-side platform (DSP): helps buyers evaluate and bid for inventory.
• Supply-side platform (SSP): helps publishers sell inventory.
• Ad exchange: facilitates transactions between demand and supply.
• Publisher: owns or operates the media environment.
• Verification and measurement providers: assess viewability, invalid traffic, suitability, reach, or outcomes.

When a user loads a page or app, an advertising opportunity may be passed to an exchange. Eligible buyers receive information permitted by the platform and privacy rules. The DSP estimates the value of the opportunity and submits a bid. The exchange selects a winning bid according to its auction and eligibility rules, and the creative is returned to the placement.

Auctions and price formation

In a simplified first-price auction, the winning bidder pays its submitted bid, subject to the platform’s rules. In a simplified second-price auction, the winner pays a price related to the next-highest eligible bid plus a specified increment. Real platforms may use floors, quality adjustments, proprietary ranking, fees, pacing, and hybrid rules. Therefore, students should not assume that every digital auction is a pure textbook first-price or second-price auction.

A campaign may optimise for a target CPM, click, conversion, completed view, or value. The bid is a signal of the buyer’s willingness to pay for an opportunity, not a guarantee that the user will notice or respond. Quality, relevance, predicted action probability, and budget pacing may affect delivery.

4. Budgeting and media metrics

A planner should select metrics that correspond to the objective and the stage of the customer journey.

The CPM formula assumes impressions are counted as the denominator and cost is in the relevant currency. CTR does not measure sales quality. CPA depends on the conversion definition, attribution window, and tracking integrity. ROAS is not profit: it excludes or may fail to reflect product cost, fulfilment, overhead, returns, and the incremental nature of sales.

Worked example: evaluating a campaign

An online retailer spends £12,000 and receives 4,000,000 impressions, 20,000 clicks, and 800 purchases. Attributed revenue is £36,000.

CPM = £12,000 ÷ 4,000,000 × 1,000 = £3.00.

CTR = 20,000 ÷ 4,000,000 × 100 = 0.5%.

Conversion rate = 800 ÷ 20,000 × 100 = 4%.

CPA = £12,000 ÷ 800 = £15.

ROAS = £36,000 ÷ £12,000 = 3.0, or £3 of attributed revenue per £1 of advertising cost.

These figures describe delivery and attributed outcomes. They do not establish causality. Some purchasers may have bought without seeing the advertisement, and some conversions may be credited to multiple channels. Incrementality testing, such as a properly designed holdout or geo-experiment, is required for stronger causal inference.

Reach planning example

Suppose a campaign produces 900,000 impressions and reaches 300,000 distinct people in a target population of 1,000,000. Average frequency is:

The reach rate is:

This is a 30% reach rate at an average frequency of three. It does not show the distribution: some people may have received one exposure and others ten. Frequency distribution and platform overlap should be examined where available.

5. Creative effectiveness and optimisation

Media and creative are interdependent. A highly relevant audience may not respond to a weak message, and an excellent advertisement may fail if delivered in an unsuitable environment. Effective creative typically provides a clear brand cue, a meaningful proposition, visual or verbal salience, a credible reason to believe, and an appropriate call to action.

Creative evaluation should consider:

1. Attention: Is the message noticed within the time available?
2. Brand linkage: Can viewers identify the advertised brand?
3. Comprehension: Is the intended meaning understood?
4. Persuasion: Does the message change preference or intention?
5. Action: Does it generate a useful response?
6. Experience: Is it accessible, respectful, and non-intrusive?

Testing may use controlled A/B experiments, randomized platform experiments, brand-lift studies, attention research, usability testing, or post-exposure surveys. A/B tests should change one major element where feasible, maintain comparable audiences and delivery periods, define a primary outcome in advance, and avoid declaring a winner from a very small or unstable sample. Platform optimisation can favour immediate clicks even when the business objective is profit or long-term brand growth; this is a form of objective mismatch.

Dynamic creative optimisation can combine images, headlines, offers, and calls to action algorithmically. It can improve relevance, but governance is required to prevent inappropriate combinations, inconsistent claims, discriminatory delivery, or unapproved use of personal data. Generative AI can assist with ideation, resizing, translation, and variant production, but human review remains essential for factual accuracy, intellectual property, accessibility, cultural sensitivity, and brand consistency.

6. Brand safety, suitability, privacy, and advertising fraud

Brand safety and suitability

Brand safety concerns preventing advertising from appearing in environments associated with illegal, hateful, extremist, violent, deceptive, or otherwise harmful content. Brand suitability is more context-dependent: an environment may be legally safe but unsuitable for a particular brand, such as content involving tragedy, controversy, or mature themes.

Controls include keyword and topic exclusions, allowlists, blocklists, pre-bid verification, post-bid monitoring, placement review, contractual requirements, and escalation procedures. Overly broad blocking can reduce reach, exclude legitimate journalism, or create discriminatory effects. Governance should therefore specify risk tolerance rather than rely on a universal list of forbidden words.

Privacy and responsible targeting

Digital advertising must respect applicable privacy and data-protection law, platform policies, and consumer expectations. Core principles include transparency, purpose limitation, data minimisation, security, retention control, lawful processing or valid consent where required, and meaningful choices. Sensitive characteristics should not be used irresponsibly for targeting or exclusion. Children and vulnerable audiences require heightened protection.

A privacy-responsible media plan documents data sources, legal basis, consent signals, processors, retention periods, access controls, and deletion procedures. Advertisers should not assume that pseudonymous identifiers eliminate privacy risk.

Advertising fraud and invalid traffic

Advertising fraud is deliberate activity intended to generate illegitimate advertising revenue or distort performance. Examples include bot traffic, non-human impressions, fake clicks, click farms, domain spoofing, ad stacking, pixel stuffing, and malicious or hidden placements. Invalid traffic is a broader measurement category that may include both deliberate fraud and non-malicious activity that does not represent a valid advertising opportunity.

Warning indicators include unusually high activity with no meaningful engagement, impossible geographic or device patterns, extreme placement concentration, rapid repeated clicks, suspiciously low dwell time, and conversions that cannot be independently verified. No single indicator proves fraud. Verification services, supply-path analysis, ads.txt or equivalent publisher-authorisation mechanisms, traffic-quality monitoring, contractual audit rights, and post-campaign reconciliation can reduce risk.

7. Attribution, incrementality, and measurement design

Attribution assigns credit for a conversion to one or more contacts. Common rules include last-click, first-click, linear, time-decay, position-based, and algorithmic attribution. Each rule embeds assumptions. Last-click is simple but tends to favour lower-funnel channels; first-click favours discovery; multi-touch models require reliable identity and sufficient data; algorithmic models may be difficult to explain.

Attribution is not the same as causal impact. If existing customers who were already likely to purchase receive retargeting ads, the platform may claim credit without creating much additional demand. Incrementality asks a different question: what happened because advertising occurred compared with what would have happened otherwise?

A basic controlled experiment compares an exposed or treatment group with a comparable holdout or control group. If the treatment group has conversion rate pₜ and the control group has conversion rate p꜀, an estimate of absolute incremental lift is:

Relative lift, when p꜀ is greater than zero, is:

The estimate assumes valid treatment assignment, limited contamination between groups, stable measurement, and a sufficiently large sample. Statistical uncertainty should be reported rather than treating any observed difference as proof of effect.

8. Illustrative case: launching an ethical personal-care brand

Background

The fictional company HarborKind is launching a refillable personal-care product. Its objective is to generate profitable online sales while building awareness among adults aged 25–44 who value convenience and reduced packaging. The initial budget is £60,000 for six weeks.

Problem

The team is tempted to allocate the entire budget to retargeting because retargeting produces the lowest reported CPA. However, the audience pool is small, frequency is rising, and many retargeted visitors may already have intended to purchase. The brand also wants to avoid advertising beside distressing content and does not want to rely on opaque third-party audience profiles.

Analysis

The team defines three roles. Video and selected connected-television inventory will introduce the product and demonstrate the refill process. Contextual display and native content will explain the environmental and convenience benefits. Retail media and paid search will capture high-intent shoppers near purchase. Retargeting will be limited by frequency and used as a reminder rather than the central growth engine.

A first-party measurement design is established using consented website analytics, transaction data, platform reporting, and an independent campaign dashboard. The team agrees that platform-reported conversions are diagnostic rather than the sole basis for budget decisions. Brand-suitable inventory categories, exclusion rules, authorised supply paths, viewability standards, and invalid-traffic monitoring are specified before launch.

Approach and solution

The budget is provisionally divided as follows:

The team develops three creative territories and tests them with consistent audience definitions. The video opens with the refill action and displays the brand early. Native content clearly labels sponsorship and links to evidence about packaging claims. Product ads use accurate prices and stock information. Frequency caps are applied, and audiences are excluded after purchase unless a legitimate replenishment message is intended.

For the retargeting test, a holdout group is used where technically and legally feasible. The team compares incremental purchases and contribution margin, not only attributed revenue. It also reviews reach, frequency, viewability, invalid traffic, placement suitability, CTR, conversion rate, CPA, and new-customer proportion each week.

Outcome

In the first two weeks, retargeting shows the lowest platform CPA but little incremental lift over the holdout. Video has a higher apparent CPA when judged by last-click attribution, yet it increases branded search and assisted visits. Contextual native placements produce fewer clicks than broad display but stronger time-on-content and higher post-exposure consideration. Retail media generates efficient sales but begins to reach the same shoppers repeatedly.

The team shifts part of the retargeting budget to contextual video and reduces bids on oversaturated retail placements. It retains a small retargeting investment because the holdout indicates a modest incremental effect for recent product-page visitors. The final evaluation combines sales, contribution margin, incremental lift, reach quality, brand-safety compliance, and learning value.

Learning points

The case demonstrates that the cheapest reported CPA is not necessarily the best investment. Channel roles should reflect the customer journey, and media decisions should consider incremental outcomes, not merely platform credit. Reach and frequency must be managed together; creative must be adapted to format; and privacy, suitability, fraud prevention, and disclosure are part of performance management rather than administrative afterthoughts.

9. Contemporary developments and professional practice

Several developments are reshaping digital media management:

• Privacy-enhancing measurement: Aggregation, clean rooms, modeled conversions, and first-party data partnerships are increasingly important as individual-level tracking becomes more restricted.
• Retail media expansion: Retailers are extending advertising from on-site search to off-site display, video, in-store screens, and data partnerships.
• Commerce and creator integration: Social content, creator recommendations, livestream commerce, and shoppable video increasingly connect persuasion with transaction.
• Connected television convergence: Television-style content and digital buying are converging, but deduplicated reach and frequency remain difficult.
• Automation and AI: Automated bidding, forecasting, creative variation, and anomaly detection improve efficiency but can reproduce bias, obscure decisions, or optimise the wrong objective.
• Supply-path optimisation: Buyers are reducing unnecessary intermediaries, improving transparency, and seeking more direct relationships with quality publishers.

Employability in digital media requires more than platform familiarity. Graduates should be able to write a clear brief, construct a media plan, interrogate platform claims, calculate metrics, present a budget rationale, identify measurement limitations, communicate with creative and analytics teams, and document ethical decisions. Spreadsheet literacy, experimentation, dashboard interpretation, procurement awareness, and concise stakeholder communication are especially valuable.

Best-practice checklist

Before launch, a media manager should verify that:

1. The objective and primary success measure are explicit.
2. The audience definition is justified and privacy-compliant.
3. Each channel has a documented role.
4. Budget, bids, pacing, reach, frequency, and exclusions are specified.
5. Creative is adapted to format, accessibility needs, and user context.
6. Brand-safety, suitability, fraud, and supply-chain controls are active.
7. Tracking, consent, attribution, and incrementality plans are documented.
8. Tests have a defined hypothesis, comparison, duration, and decision rule.
9. Platform-reported results are reconciled with independent or first-party evidence.
10. Optimisation considers long-term brand and customer value, not only immediate clicks.

Challenges and limitations

Digital media measurement remains fragmented. Different platforms use different definitions of an impression, view, conversion, and user. Cross-device identity can be incomplete, consent can reduce observable data, and modeled results can be difficult to audit. Reach estimates may be deduplicated poorly, while frequency may be understated when a person uses several devices.

Auction transparency is also limited. Advertisers may not know the full sequence of fees, quality adjustments, or competing bids. Automated systems can optimise toward measurable proxies rather than strategic outcomes. Narrow targeting may reduce diversity and create unfair exclusion, while broad targeting may waste budget. Finally, digital advertising can contribute to surveillance, misinformation funding, environmental costs from data processing, and manipulative design. Responsible management therefore requires both commercial effectiveness and public accountability.

Chapter Summary

Digital advertising and media management connect objectives, audiences, channels, creative, budgets, measurement, and governance. Display, video, native, retail media, search, social, and connected television serve different roles and should be evaluated within an integrated plan. Reach describes unique exposure, frequency describes average repetition, and metrics such as CPM, CTR, CPA, and ROAS provide useful but incomplete evidence.

Programmatic advertising automates the buying and selling of inventory through DSPs, SSPs, exchanges, and auction mechanisms. Because real auctions include quality, fee, pacing, and eligibility rules, simplified first-price and second-price models must not be treated as universal descriptions. Effective management combines audience relevance, creative quality, experimentation, incrementality analysis, and appropriate attribution.

Brand safety, suitability, privacy, disclosure, accessibility, and fraud prevention are central managerial responsibilities. The strongest media plans do not simply maximise reported conversions; they seek profitable and incremental outcomes, protect audiences and brands, and create reliable organisational learning.

Key Terms

Advertising fraud: Deliberate activity intended to generate illegitimate advertising revenue or distort performance.

Attribution: A rule or model for assigning conversion credit to marketing contacts.

Brand safety: Controls designed to prevent advertising from appearing in harmful or unacceptable environments.

Brand suitability: The degree to which a media environment fits a particular brand’s risk tolerance and identity.

Connected television: Television-like content delivered through internet-connected devices.

Contextual targeting: Selecting placements according to the content or environment rather than an inferred individual profile.

CPM: Cost per thousand impressions.

DSP: Demand-side platform used by buyers to evaluate and purchase digital inventory.

First-party data: Data collected directly by an organisation from its own customers or audiences.

Frequency: Average number of exposures among reached individuals.

Incrementality: The additional outcome caused by advertising compared with what would have occurred without it.

Invalid traffic: Traffic that does not represent a valid advertising opportunity, whether fraudulent or otherwise unsuitable for measurement.

Native advertising: Paid content designed to resemble the form and function of its surrounding media while being disclosed as advertising.

Programmatic advertising: Software-enabled buying and delivery of digital advertising inventory.

Reach: Number or proportion of distinct individuals exposed at least once.

Retail media: Advertising sold through or using a retailer’s digital, physical, or shopper-data assets.

ROAS: Attributed advertising revenue divided by advertising cost.

SSP: Supply-side platform used by publishers to manage and sell inventory.

Viewability: The extent to which an impression meets a defined opportunity-to-see standard.

Supply-path optimisation: Evaluation and reduction of inefficient or low-quality routes through the advertising supply chain.

Review Questions

Short Answer Questions

1. Distinguish between paid, owned, and earned media.
2. Define reach and frequency, and explain how they are related.
3. What is the role of a demand-side platform in programmatic advertising?
4. Why is ROAS not equivalent to profit?
5. Give two examples of controls used to reduce advertising fraud.

Long Answer Questions

1. Explain how a digital media plan should connect business objectives, audience insight, channel roles, creative, budget, and measurement.
2. Compare display, video, native, and retail media in terms of strengths, limitations, and appropriate uses.
3. Explain programmatic advertising and discuss why real-world auctions may differ from simplified first-price and second-price models.
4. Discuss the relationship between creative effectiveness, media context, and campaign performance.
5. Evaluate the ethical and managerial importance of privacy, brand safety, suitability, and disclosure.

Analytical / Application Questions

1. A campaign costs £18,000, generates 3,000,000 impressions, 15,000 clicks, and 600 conversions, and produces £45,000 in attributed revenue. Calculate CPM, CTR, conversion rate, CPA, and ROAS. Explain two limitations of interpreting these figures.
2. A retailer reports that retargeting has the lowest CPA, but a holdout group shows almost the same purchase rate as the exposed group. Analyse why platform attribution may overstate performance and recommend a budget decision.
3. Design a six-week media plan for a new sustainable product with a limited budget. Specify audience approach, roles for at least three channels, reach and frequency considerations, creative tests, fraud and brand-safety controls, and an incrementality measure.

Multiple Choice Questions with Answers

Question 1

Which metric represents the average number of exposures among people reached by a campaign?

A. Reach

B. Frequency

C. CPM

D. ROAS

Correct Answer: B. Frequency

Question 2

A campaign costs £12,000 and delivers 4,000,000 impressions. What is its CPM?

A. £0.30

B. £3.00

C. £30.00

D. £300.00

Correct Answer: B. £3.00

Question 3

Which platform primarily helps advertisers evaluate and bid for digital inventory?

A. SSP

B. DSP

C. Publisher CMS

D. Payment gateway

Correct Answer: B. DSP

Question 4

Which statement best describes native advertising?

A. Advertising that is always delivered through television

B. Advertising designed to resemble the surrounding media while being disclosed as paid content

C. Advertising shown only to existing customers

D. Advertising that cannot contain a call to action

Correct Answer: B. Advertising designed to resemble the surrounding media while being disclosed as paid content

Question 5

A campaign has 20,000 clicks from 4,000,000 impressions. What is the CTR?

A. 0.05%

B. 0.5%

C. 5%

D. 50%

Correct Answer: B. 0.5%

Question 6

Why can a low retargeting CPA fail to demonstrate incremental effectiveness?

A. Retargeting never produces sales

B. Retargeted users may already have been likely to purchase

C. CPA measures impressions rather than conversions

D. Retargeting cannot be measured

Correct Answer: B. Retargeted users may already have been likely to purchase

Question 7

Which practice most directly supports brand suitability?

A. Removing all frequency caps

B. Using placement exclusions and context-specific risk rules

C. Maximising the number of intermediaries

D. Optimising only for clicks

Correct Answer: B. Using placement exclusions and context-specific risk rules

Question 8

What is the principal purpose of a holdout group in advertising evaluation?

A. To increase the campaign’s frequency automatically

B. To estimate what would have happened without advertising

C. To guarantee every user sees an advertisement

D. To calculate a publisher’s CPM

Correct Answer: B. To estimate what would have happened without advertising

Question 9

Which statement about ROAS is correct?

A. It is always equal to profit margin

B. It is attributed advertising revenue divided by advertising cost

C. It measures unique reach only

D. It proves that advertising caused every reported sale

Correct Answer: B. It is attributed advertising revenue divided by advertising cost

Question 10

Which is an example of first-party data?

A. A customer’s consented purchase history held by the retailer

B. An anonymous list purchased from an unrelated broker

C. A competitor’s private CRM file

D. A random bot-generated clickstream

Correct Answer: A. A customer’s consented purchase history held by the retailer